Alcohol use in adolescence is a leading contributor to accidental injuries, mental health problems, and mortality. In addition, adolescent drinking is commonly co-morbid with other risky behaviors, including smoking, illicit drug use, and delinquency. Despite concerted efforts at prevention, drinking and other risky behaviors are still widely prevalent in adolescence. Emerging research in neuroscience has suggested a novel biological mechanism for understanding developmental differences in risk-taking. Recent evidence suggests sub-cortical brain regions, which underlie automatic responses to emotions and rewards (hot influences on decision making), mature more quickly than cortical brain regions, which underlie planning and inhibition (cold deliberative influences on decision-making) and mature slowly through the mid-20s. Consistent with this model of adolescent brain development, our previous research found that performance on behavioral measures of cognitive control improves from early adolescence through early adulthood, whereas behavioral measures of reward sensitivity show a U-shaped pattern of development, with the highest reward sensitivity seen in middle adolescence. Thus middle adolescence (ages 15-16) is a particularly vulnerable period: When faced with a emotionally charged, highly novel, potentially pleasurable situation - like the offer of a drink - responding in the hot affective system is likly to overwhelm an adolescent's cold deliberative control. This research will extend our understanding of how adolescent decision-making influences alcohol use and other risky behaviors using a behavioral genetics design. The project focuses on two influences on adolescent decision-making, which will be measured using a previously validated battery of survey measures and objective cognitive tests: (1) reward sensitivity (hot responses to positive, rewarding, or novel stimuli); and (2) cognitive control (cold responses that require planning and inhibition). The project will use a novel, ethnically-diverse, population- based sample of 200 adolescent twin pairs, ages 15-16 (the age group in which the disjunction between reward sensitivity and cognitive control has been found to be the highest). Alcohol use, illicit drug use, delinquency, and fighting will be measured using twin self-report, parent report and official school record data on disciplinary infractions. This multi-method approach will address the following specific questions: (1) Do individual differences in cognitive control and reward sensitivity predict individual propensities for alcohol use and other risk-taking behaviors? (2) Are cognitive control and reward sensitivity endophenotypes for genetic influence on alcohol use and other risk-taking behaviors? and (3) How do family environments moderate the impact of high reward sensitivity on risk-taking behaviors? Data will be analyzed using sophisticated quantitative methods, including analyses of gene-environment interaction (GxE). The proposed research will thus integrate neuroscientific theories of adolescent brain development with behavioral genetic research methods, in order to identify novel specific endophenotypes for adolescent risk-taking.